FastAIRag 0.1.0 — Unified, Zero-Bloat RAG Pipeline Client for Java

---

💡 Connect local vector storage with generative AI models — Minimalist retrieval pipeline orchestrating chunking, embeddings, and context insertion.

FastAIRag is a lightweight, framework-agnostic RAG engine designed to feed relevant context into local or cloud models with zero framework bloat. It orchestrates file chunking, embedding generation, index mapping, and query formatting, giving local models (like 1B-3B LLMs) the domain knowledge they need to avoid hallucinations.

---

Key Features

• 📂 Modular Chunking — Ingest entire file systems or documents, parsing them into uniform sliding chunks.
• 🔌 Model Agnostic — Easily integrate local embedders (like LM Studio or Ollama) using the single-method EmbeddingProvider.
• ⚡ Integrated Search — Automatic routing and retrieval using the high-performance FastAIVectorDB engine.
• 🤖 Direct Pipe Integration — Seamlessly feed retrieved context as system prompts directly into your FastAI connection.

---

Quick Start

import fastai.AI;
import fastai.FastAI;
import fastairag.*;

// 1. Define how your app obtains embeddings
EmbeddingProvider embedder = text -> myEmbeddingAPI(text);

// 2. Build the Document Store and index files
RagStore store = FastAIRag.store(embedder);
store.addDirectory(Path.of("./docs"), 512, 128);

// 3. Attach AI to generate context-supported answers
AI llm = FastAI.connect("ollama:llama3.2");
RagPipeline pipeline = FastAIRag.pipeline(llm, store);

String response = pipeline.ask("How do I compile the native package?");
System.out.println(response);

---

Installation

Maven (JitPack)

Add JitPack repository and dependency configuration to your pom.xml:

<repositories>
    <repository>
        <id>jitpack.io</id>
        <url>https://jitpack.io</url>
    </repository>
</repositories>

<dependencies>
    <dependency>
        <groupId>com.github.andrestubbe</groupId>
        <artifactId>FastAIRag</artifactId>
        <version>0.1.0</version>
    </dependency>
    <dependency>
        <groupId>com.github.andrestubbe</groupId>
        <artifactId>FastAIVectorDB</artifactId>
        <version>0.1.0</version>
    </dependency>
</dependencies>

---

Technical Architecture

Offline Path:
[Raw Docs] ---> [Chunker] ---> [EmbeddingProvider] ---> [FastVectorDB]

Online Path:
[User Query] ---> [EmbeddingProvider] ---> [FastVectorDB Query] ---> [Context]
                                                                        |
                                                                        v
[Response] <--- [FastAI LLM Inference] <--- [Injected System Prompt] <--+

---

Part of the FastJava Ecosystem — Making the JVM faster. Small package. Maximum speed. Zero bloat. 🚀📋